Inorg. Chem. 1988, 27, 81-85
81
Contribution from the Departments of Chemistry, University of Miami, Coral Gables, Florida 33 124, and Florida International University, Miami, Florida 33 199
Heat of Reaction of (Norbornadiene)molybdenum Tetracarbonyl with Monodentate and Bidentate Ligands. Solution Thermochemical Study of Ligand Substitution in the Complexes cis -L,Mo(CO), Shakti L. Mukerjee,t Steven P. Nolan,t Carl D. Hoff,*t and Ramon Lopez de la Vega$ Received June 4, 1987 The enthalpy of reaction of (NBD)Mo((30), (NBD = norbomadiene) with a number of monodentate and bidentate ligands forming cis-L,Mo(CO), has been measured at 30 OC in T H F solution. The heats of reaction span a range of 33 kcal/mol. The order of stability for monodentate ligands is PCl, < As(C6H5)3 < py < P(C6H5), < AsEt, < P(OC6H5), < P(C6HS),(CH3)< CO < P(C~HS)(CH,)Z < P(OCH3), < P(n-Bu), < (C6H,,)NC < PEt, < PMe3. The series of chelating bidentate phosphines RIP(CH2),PR2 (n = 1-4, R = C6H5;n = 1, 2, R = CHI) and several related ligands were investigated. The chelating ring systems in the metallacycles show strain energies of about 8 kcal/mol for four-membered rings. The mixed ligand (C6H5)zPCH2CH2A s ( C ~ H shows ~ ) ~ a heat of binding midway between the heats of binding of (C6H5)2PCHICH2P(C6H5)2 and (C6H5),AsCH2CH2A~(C6H5)2, implying group additivity in this system. The complex (phen)Mo(CO), is some 5 kcal/mol more stable than (bpy)Mo(CO),, presumably due to conformational effects in the free ligand. The ligand 1,5-cyclooctadiene forms a complex 2 kcal/mol less stable than that of norbornadiene. The influences of steric and electronic factors in determining the Mo-L bond strength are discussed.
Introduction Chelating ditertiary phosphines have been widely used as ligands in organometallic chemistry. Kinetic, catalytic, and structural studies have been made of these complexes.’ There are no thermodynamic d a t a regarding strain energy or heats of binding for these ligands. Several calorimetric studies for monodentate phosphines have been reported: Manzer and Tolman investigated the solution thermochemistry of square-planar platinum(I1) complexes, shown in eq 1, for a series of phosphine ligands.2 A trans-(CH3PtQ2(THF)]+
+ PR3
-
trans-[CH3PtQ2(PR3)]+
+ THF
(1)
Q = PMe2Ph
similar series of ligands was then investigated3 for the Ni(0) complexes shown in eq 2. These d a t a were interpreted in terms Ni(COD)2
+ 4PR3
-
Ni(PR&
+ 2COD
(2)
of the steric a n d electronic parameters of t h e phosphine ligand by using t h e cone angle as a measure of steric bulk and the A , vibrational frequency in Ni(C0)3L as a measure of ligand basi~ity.~ Recently, we reported a solution thermochemical study of ligand substitution in the c o m p l e x e s f ~ c - L ~ M o ( C 0for ) ~ phosphine5 and other ligands:6 (arene)Mo(C0)3
+ 3 L - + f ~ c - L ~ M o ( C 0 )+~ arene
(3)
The present study extends our work on solution thermochemistry to a series of cis-L2Mo(CO), complexes. In addition to measuring heats of binding of monodentate ligands, we report the first extensive data on the effect of chelate ring size on complex stability for bidentate phosphines. Experimental Section All manipulations were carried out under an argon atmosphere by using standard Schlenk tube techniques or in a Vacuum Atmospheres glovebox with less than 1 ppm of oxygen and water. Tetrahydrofuran was purified by distillation from calcium hydride to a second flask, where it was distilled from sodium benzophenone ketyl under argon into flame-dried glassware. Other reagents were obtained from commercial sources and purified by standard procedures. Organomolybdenum compounds were prepared by using procedures described in the literature’ or slight modifications thereof. Complexes used in calorimetry were subjected to at least two purification steps involving either recrystallization or high-vacuum sublimation. Only materials that were pure as determined by IR and NMR spectroscopy and were of high crystalline quality
7 University f
of Miami. Florida International University.
0020-1669/88/1327-0081$01.50/0
were used in the calorimetric measurements. If a new batch of material was used, measurements of heats of solution and heats of reaction were repeated to ensure that the material was not only chemically pure but had the same heat content. Infrared spectra were obtained by using a Perkin-Elmer 267 or 501 spectrophotometerin 0.1 mm KBr solution cells. NMR spectra were obtained on a Varian FT-80A spectrometer. All calorimetric measurements were made at 30 OC by using the Calvet calorimeter (Setaram C-80) described previ~usly.~ Reaction of (NBD)Mo(CO)~with 1,2-Bis(diphenylphosphino)ethane (diphos). In a typical procedure the mixing cells of the calorimeter were dried in an oven at 150 OC for 2 h and then taken into the glovebox. The lower chamber was filled with approximately 0.15 g of (NBD)Mo(CO),, closed, and sealed with 1.5 mL of mercury. Four milliliters of a 25% solution of diphos in THF was added to the cell. It was then closed, taken out of the glovebox, placed in the sample compartment of the calorimeter, and allowed to equilibrate at 30 OC for 2 h. After the temperature reached a stable base line, the reaction was initiated by inverting the calorimeter. After return to the base line, the trace was integrated and the enthalpy of reaction calculated. At the end of the run, IR spectra confirmed that reaction was complete and produced (diphos)Mo(CO), as the only organometallic product. The value of -20.5 f 0.3 kcal/mol represents the average of five different measurements. This value refers to reaction of the crystalline solid with diphos producing the diphos complex in solution. The enthalpy of solution of the crystalline solid (+3.8 kcal/mol) must be subtracted from the measured heat of reaction -20.5 kcal/mol) in order to obtain the enthalpy of reaction with all species in solution (-24.3 kcal/mol). Reaction of (NBD)Mo(CO), with CO: High-pressure Calorimetry. In a typical reaction the high-pressurecells of the Calvet calorimeter were taken into the glovebox and loaded under an argon atmosphere. The reference cell was loaded with 4.0 mL of THF and the sample cell loaded with 0.1 g of (NBD)Mo(CO), in 4.0 mL of THF. Matheson research grade CO containing less than 1 ppm of O2 was transferred to an autoclave containing a T H F solution of sodium benzophenone and thermostated at 30 OC. Gas lines leading to the calorimeter were also thermostated at 30 OC. After the cells had equilibrated, both were pressurized with CO at 500 psi. The enthalpy of reaction, -22.6 (0.8) kcal/mol, is the average of six determinations. The infrared spectrum showed M o ( C O ) ~as the sole metal-containing product. Reactions on samples of more than 0.1 g of (NBD)Mo(CO), went to completion more slowly, since it appears that the carbonylation reactions are inhibited by the norbornadiene produced in the reaction. (1) Pignolet, L. H., Ed. Homogeneous Catalysis with Metal Phosphine Complexes; Plenum: New York, 1983. (2) Manzer, L. E.; Tolman, C. A. J . Am. Chem. SOC.1975, 97, 1955. (3) (a) Tolman, C. A.; Seidel, W. C.; Gosser, L. W. J . Am. Chem. SOC. 1974, 96, 53. (b) Tolman, C. A.; Reuter, D. N.; Seidel, R. W. J. Organomet. Chem. 1976, 117, C30. (4) Tolman, C. A,; Chem. Rev. 1977, 77, 313 and references therein. (5) Nolan, S. P.; Hoff, C. D. J . Organomet. Chem. 1985, 290, 365. (6) (a) Hoff, C. D. J . Organomet. Chem. 1983, 246, C53. (b) Hoff, C. D. J . Organomet. Chem. 1985, 282, 215. (c) Nolan, S. P.; Lopez de la Vega, R.; Hoff, C. D. Organometallics 1986, 5, 2529. (d) Nolan, S. P.; Hoff, C. D.; Landrum, J. T. J . Organomef.Chem. 1985,282,365. ( 7 ) Wilkinson, G., Stone, F. G. A,, Abel, E. W., Eds.; Comprehensiue Organometallic Chemistry; Pergamon: New York, 1982; Vol. 3.
0 1988 American Chemical Society
82 Inorganic Chemistry, Vol. 27, No. 1, 1988
Mukerjee et al.
Table I. Enthalpies of Ligand Exchange (kcal/mol) (NBD)Mo(CO), + 2L L,Mo(CO), + NBD +
and Bond Strength Estimates (kcal/mol) for Monodentate Ligands bond ligand complex -AH strength norbornadiene 0 26.7" 7.4 (0.2) 30.4 phosphorus trichloride 10.5 (0.3) 32.0 triphen ylarsine 12.1 (0.1) 32.8 pyridine 17.1 (0.1) 35.2 triphen ylphosphine 19.7 (0.2) 36.6 triethylarsine 21.9 (0.1) 37.7 triphenyl phosphite 22.2 (0.2) 37.8 diphenylmethylphosphine 22.6 (0.8) 38 f 3 carbon monoxide 26.8 (0.4) 40.1 phenyldimethylphosphine 27.5 (0.1) 40.5 trimethyl phosphite 30.1 (0.1) 41.8 tri-n- butylphosphine 30.1 (0.1) 41.8 cyclohexyl isocyanide 32.1 (0.1) 42.8 triethylphosphine 32.9 (0.1) 43.2 trimethyl phosphine nPer olefin.
Results and Discussion The goal of this work was investigation of the enthalpy of ligand exchange and evaluation of strain energies in the complexes cis-L,Mo(CO),. The entryway to the thermochemistry in this system is reaction of (NBD)Mo(CO), with excess ligand as illustrated in reaction 4 for trimethylphosphine. These reactions (NBD)Mo(CO), 2PMe3 (PMe3)2Mo(CO)4+ NBD (4) AH = -32.9 (0.1) kcal/mol
+
-
are rapid and quantitative at room temperature. The initially formed cis products are known to undergo thermal isomerization to yield cis/trans mixtures, but this reaction is relatively slow.s Infrared spectra run immediately after calorimetry showed no detectable amount of trans product. In addition, it appears that there is little difference in energy between the cis and trans isomers for small phosphines.s All data in this paper refer to the cis isomer. Thermochemical results for monodentate ligands are collected in Table I and provide a basis for calculating enthalpies of ligand substitution. For example, reaction with trimethyl phosphite proceeds as shown in eq 5 . Subtraction of eq 5 from eq 4 leads (NBD)Mo(CO), 2P(OMe), [P(OMe),12Mo(CO), + NBD ( 5 )
+
-
AH = -27.5 (0.1) kcal/mol directly to the enthalpy of the substitution reaction shown in eq 6, since all other terms cancel. In a similar fashion the enthalpy [P(OMe),],Mo(CO), + 2PMe3 (PMe,),Mo(CO), + 2P(OMe), (6)
-
AH = -5.4 (0.2) kcal/mol of interconversion of any of the complexes can be calculated. Bond Strength Estimates. The bond strength estimates shown in Table I refer to solution-based values. The heat of reaction with carbon monoxide, shown in eq 7, provides the platform for (NBD)Mo(CO), + 2CO Mo(CO), + NBD (7) --L
AH = -22.6 (0.8) kcal/mol assigning absolute values to the bond strengths shown. Recent laser photolysis studies of M o ( C O ) ~ ,in the gas phase, led to estimation of the first bond dissociation energy as 40.5 kcal/moL9
Table 11. Heats of Reaction, Cone Angles, and Electronic Parameters for Selected Phosphines (NBD)Mo(CO), + 2PR3 (PR,),Mo(CO), + NBD
-
phosphine or phosphite
-AHu 17.1 21.9 22.2 26.8 27.5 30.1 32.1 32.9
ob
VC
145 128 136 122 107 132 132 118
2068.9 2085.3 2067.0 2065.3 2079.5 2060.3 2061.7 2064.1
Reaction enthalpies are reported in kcal/mol. Cone angles are in degrees and are taken from ref 4. 'Electronic parameters are in cm-' and are taken from ref 4.
Table 111. Heats of Reaction and X-ray Crystallographic Data for Selected Phosphines Mo-PC ~ ( M o - P ) ~ @e phosphine -AH" PMe, 32.9 0 2.522 0 97.54 PEt, 32.1 0.8 2.544 0.022 100.27 P-n-Bu3 30.1 2.8 0.030 99.29 2.552 PMe2Ph 26.8 6.1 2.529 0.007 94.78 PPh,Me 22.2 10.7 2.555 0.033 92.52 PPh, 17.1 15.8 2.577 0.055 104.62 "All values reported are in kcal/mol. bRelative stabilities (in kcal/mol) with PMe, as reference. CBonddistances (in angstroms) are taken from ref 8. "Bond length difference with PMe, (in angstroms). 'P-W-P angle (in degrees).
The average bond dissociation energy for all six carbonyl ligands was determined as 36 kcal/mol.1° The value 38 f 3 kcal/mol is adopted here and leads directly to assignment of the bond strength estimates in Table I. Due to uncertainties in the transferability of bond strength data from the gas phase to the solution phase, we consider these bond strength estimates accurate to within 5 kcal/mol. The enthalpies of substitution in solution are known to greater accuracy (typically to 0.5 kcal/mol), and these are used directly to estimate complex stability. Enthalpy of Phosphine Substitutionand Its Relation to Steric and Electronic Properties of the Phosphine Ligand. The steric and electronic properties of phosphine ligands are readily described in terms of the cone angle and electronic factor, originally proposed by T01man.~The enthalpy of reaction 4 was shownS to fit eq 8 AH = A,
+ A,O + A2v
(8)
with a correlation coefficient of 0.95 (Ao = -613.00, A , = 0.36, A2 = 0.24). Attempts to fit the enthalpy of reaction for the monodentate phosphines listed in Table I1 (along with cone angles and electronic parameters) to eq 8 yielded R = 0.86 (Ao = -993.13, A , = 0.36, and A2 = 0.445). Despite the low correlation coefficient, useful information may be obtained from this type of analysis. The ratio of A , / A 2can be taken as a measure of the relative importance of steric versus electronic factors. It would be expected that steric factors would become less important in going from the L3Mo(CO), system to the L,Mo(CO), system. The change in the A 1 / A 2ratio shows a decreasing dependence on steric factors in going from L,Mo(CO), (1.42) to L2Mo(CO), (0.81). Work in progress on the thermochemistry of L M O ( C O ) ~ complexes should show a continued decrease in the A 1 / A 2ratio. Enthalpy of PhosphineSubstitution and Its Relation to Structural Features of the Complexes. The analysis of ligand behavior in terms of the cone angle and electronic factors is simple and is perhaps the best way to first view data of this kind. For many of the complexes studied X-ray structures have been reported*
(8) (a) Darensbourg, D. Inorg. Chem. 1979, 18, 14. (b) Darensbourg, D. J.; Graves, A. H. Inorg. Chem. 1979, 18, 1257. (c) Cotton, F. A,; Darensbourg, D. J.; Klein, S.; Kolthammer, B. W. S . Inorg. Chem. 1982, 21, 2661. (d) Cotton, F. A,; Darensbourg, D. J.; Klein, S.; Kolthammer, B. W. S. Inorg. Chem. 1982, 21, 294.
(9) Lewis, K. E.; Golden, D. M.; Smith, G . P. J . Am. Chem. SOC.1984,106, 3905. (10) Barnes, D. S.; Pilcher, G.; Pittman, D. A,; Skinner, H. A,; Todd, D.; Virmani, Y . J . Less-Common Met. 1974, 36, 177.
Inorganic Chemistry, Vol. 27, No. 1, 1988 83
Heat of Reaction of (NBD)Mo(CO), and this allows a second comparison. It must be kept in mind, of course, that thermochemical data are generated in solution and that solid-state structures may show significant differences from those in solution. Molybdenum-phosphorus bond lengths, P-Mo-P angles, and heats of reaction are reported in Table I11 for complexes with known crystal structures. Relative heats of displacement and bond lengths (with ( P M e 3 ) 2 M ~ ( C 0 )as 4 reference) are also shown. There is no obvious correlation between the complex stability and the P-Mo-P bond angle. There is a rough correlation with the Mo-P bond length. It should be noted that sequential replacement of methyl groups by phenyl results in a steady decrease in the enthalpy of reaction. There is a smooth transition from PMe, to PMe2Ph to PMePhz to PPh, in steps of about 5 kcal/mol (see Table 111). This corresponds to individual bond strength decreases on the order of 2.5 kcal/mol (since 2 mol of phosphine is bound) as methyl is replaced by phenyl. Electronic Requirements of the Metal Center. The enthalpies of substitution for the monodentate ligands are, in most cases, close to those that would be predicted from earlier work on fuc-L,Mo(CO),. For example, the enthalpy of reaction 9 is -44.4 (CHPT)Mo(CO),
+ 3P(OMe),
-
[P(OMe),],Mo(CO),
+ CHPT
Table IV. Enthalpies of Reaction and Bond Strength Estimates for Bidentate Ligands" (NBD)Mo(C0)4 + LL (LL)Mo(CO), + NBD ligand
(COD)Mo(C0)4
P
-AH
bond strength
-2.0 (0.2)
25.7
complex
(NBD)Mo(CO),
26.7
0
32.6 11.9 (0.3) 13.4 (0.3)
32.7 33.4
(Me2phen)Mo(CO)4 15.8 (0.2)
34.6
(dppm)Mo(CO),
16.0 (0.2)
34.7
(phen)Mo(CO),
17.0 (0.2)
35.2
19.3 (0.1) 22.1 (0.1)
36.4 31.8
23.0 (0.1) 24.3 (0.3) 24.5 (0.5)
38.2 38.9 39.0
(dpae)Mo(CO), (tmeda)Mo(CO),
(9)
CHPT = cycloheptatriene kcal/mol.6 This involves replacement of three Mo-olefin bonds by 3 mol of P(OMe),. An estimate for the enthalpy of reaction 5 could reasonably be made by taking two-thirds of the value for the enthalpy of reaction 9. The estimated value (-29.6 kcal/mol) is close to the experimental value (-27.5 kcal/mol). For many of the ligand-exchange reactions studied here, the enthalpies of reaction could have been predicted to fair accuracy from the earlier work on L,Mo(CO),. However, such calculations do show unexpected differences as illustrated below. In the fuc-L3Mo(C0), system, trimethyl phosphite and trimethylphosphine are nearly equal in strength. The enthalpy of reaction 10 is +1.4 (0.8) kcal/mol, implying that the phosphite fac-[P(OMe),],Mo(CO),
+
-
PhzPCH2CH2AsPh2 (Arphos)Mo(CO), P ~ ~ P C H ~ C H ~ C H ~ C(dppb)Mo(CO), HZPPh2 Ph2PCH2CH2CH2PPh2 (dppp)Mo(CO), Ph2PCH2CH2PPh2 (dppe)M~(CO), Me2PCH2PMe2 (dmpm)Mo(CO),
am
(dppbz)Mo(CO),
24.7 (0.8)
39.1
Me2PCH2CH2PMe2
(dmpe)Mo(CO),
32.4 (0.1)
42.9
FPhz
" Enthalpies and bond strengths are given in kcal/mol.
3PMe3 ( P M ~ , ) , M o ( C O ) ~ 3P(OMe), (10)
+
is a slightly stronger binding ligand. This situation is reversed in the complexes ~ i s - L ~ M o ( C 0 The ) ~ . enthalpy of reaction 6 is -5.4 (0.2 kcal/mol, indicating that trimethylphosphine is a stronger ligand in this system. Since trimethylphosphine has a smaller cone angle than trimethyl phosphite, this change cannot be ascribed to steric factors. The origin must be in the changing electronic requirements of the metal center. Addition of the highly basic PMe, becomes less favorable with successive substitution due to buildup of excess electron density at the metal center. When complete data are in for the LMo(CO), system, it may be possible to examine electronic effects in more detail. Additional Monodentate Ligands. The data in Table I include monodentate amine, arsine, and isocyanide ligands as well as carbon monoxide. The general trend observed forfac-L,Mo(CO),, amine arsine < phosphine, is retained in this series. It was shown earlier that the nature of the substituents in PR3 complexes affects complex stability. This is true for arsines as well; the AsEt, complex is 9.2 kcal/mol more stable than the corresponding AsPh, complex. The difference between PEt, and PPh, is 15 kcal/mol. As was the case for L,Mo(CO),, carbon monoxide and isocyanides are among the strongest ligands. They have bond strengths comparable to those of phosphine ligands, as would be expected from the synthetic chemistry. Bidentate Ligands. The enthalpies of reaction for bidentate ligands are listed in Table IV. Examination of the bond strength data and heats of ligand substitution for monodentate ligands revealed the general pattern in bond strengths P > As = N. In the absence of special steric features, the heats of reaction of bidentate ligands follow this pattern and are largely predictable on the basis of the choice of an appropriate monodentate ligand model. For example, the heat of reaction of 2 mol of diphenyl-
16.0
23.0
24.3
221
24.7
22.2
Figure 1. Relative stabilities (kcal/mol) for [Ph2P(CH2)J'Ph2]Mo(CO), complexes. methylphosphine (-22.2 kcal/mol) is close to that of 1,2-bis(dipheny1phosphino)ethane (dppe); -24.3 kcal/mol). The heat of reaction of bipyridine (-1 1.8 kcal/mol) is quite close to that of 2 mol of pyridine (-12.1 kcal/mol). Significant differences from expected heats of binding will be referred to as "strain energy". It should be noted that bidentate ligands should have more favorable entropies of binding leading to greater thermodynamic stability even in the absence of favorable enthalpies of ligand displacement. Chelating Phosphines and Arsines. The strain energies in cyclic organic compounds have been extensively investigated. Conventional ring strain estimates for four-, five-, and six-membered rings are on the order of 26, 6, and 0 kcal/mol." Incorporation of molybdenum into the chelate structures, shown below, yields four- to seven-membered rings with the heats of reaction as shown (1 1) Cox, J. D.; Pilcher, G. Thermochemistry of Organic and Organometallic
Compounds; Academic: New York, 1970.
84 Inorganic Chemistry, Vol. 27, No. 1, 1988
Mukerjee et al.
in Figure 1. The five-membered ring system based on either diphos or 1,2-(dipheny1phosphino)benzene (dppbz) is most stable and slightly more so than 2 mol of Ph2MeP. The six- and seven-membered ring systems are slightly less stable. Only for the four-membered system is a significant “ring strain” observed (8.3 kcal/mol). This is on the same order of magnitude as a recent estimate for thorium metallacyclobutanes (10 kcal/mol).I2 As shown in Table IV, the methyl-substituted analogues dmpm and dmpe present a similar behavior with a enthalpy difference of 7.9 kcal/mol between the four- and five-membered metallacycles. In this case, the more basic substituents lead to higher values and greater overall stability. Thus the enthalpy of replacement of dmpm by diphos, in reaction 11, is near zero. The strain energy d p p m M ~ ( C O ) ~diphos = d i p h o ~ M o ( C 0 ) ~ dppm (1 1)
+
+
in the four-membered ring of d ~ p m M o ( C 0 is ) ~nearly balanced by stronger binding of the more basic trialkylphosphine donors (dppm) versus that of the diary1 alkyl ligands (diphos). In this regard, it is also of interest that, whereas diphos was slightly more stable than 2 mol of diphenylmethylphosphine,dmpe is slightly less stable than 2 mol of trimethylphosphine. A possible explanation is that the phenyl groups of diphos may be held out of the way due to the structural demands of the chelate ring. This would lead to decreased steric repulson relative to that for diphenylmethylphosphine. For the methyl-substituted phosphines, this should be of little importance. Spectroscopic data also highlight the special position of the four-membered ring systems. The 1s3W-31Pand the 9SM0-31P coupling constantsI3 are anomalously low for the four-membered metallacycles. The structure of ( d p p m ) M ~ ( C O )shows ~ no lengthening of the Mo-P bond14 compared to that in other diphenylalkylphosphines. The P-Mo-P angle (67.3O) shows considerable distortion and does give indication of the strain present in the system. Phosphine ligands with small “bites” are known to form bridged structures as shown in reaction 12.l The entropy of reaction 12
‘PPh2 (12)
should be unfavorable and can be estimated to be on the order of 10 kcal/mol at room t e m p e r a t ~ r e . ’ ~For the molybdenum system, we estimate the enthalpy of reaction 12 to be on the order of -16 kcal/mol, clearly offsetting the unfavorable entropy term. Thus, the mononuclear complex is probably a kinetic product. Chelating arsine and mixed phosphine/arsine chelates were also studied. As expected, replacement of phosphorus by arsenic resulted in decreased stability. Bond strength estimates for five-membered rings appear to be roughly additive. The enthalpy of reaction 13 is calculated to be -1.3 (0.3) kcal/mol per mole
c: /M0(c0)4
2
Table V. Infrared Carbonyl Bands (cm-I), Enthalpies of Reaction (kcal/mol), and Bond Strength Estimates (kcal/mol) for cis-L.Mo(CO)d Complexes complex (COD)Mo(C0)4 (NBD)Mo(CO), (PCl,),Mo(CO), (Ph,As),Mo(CO), (bPY)MO(CO), (diars)Mo(CO), (PY)zMo(C0)4 (tmeda)Mo(CO), (Me2phen)Mo(CO), (dppm)Mo(CO), (phen)Mo(CO), (Ph,P)2Mo(CO), (arphos)Mo(CO), (Et,As),Mo(CO), [P(OPh),],Mo(CO), (dppb)Mo(CO), (Ph,MeP),Mo(CO), Mo(C0)6 (dppp)Mo(CO), (dppe)Mo(CO), (dmpm)Mo(CO), (dppbz)Mo(CO), (Me2PhP)2Mo(C0)4 [(MeO)$’12Mo(CO), (n-Bu,P),Mo(CO), (CyNC),Mo(CO), (Et,P),Mo(CO), (dmpe)Mo(CO), (Me,P),Mo(CO),
IR (COI bands” 2041, 2022, 2070, 2045, 2039, 2043, 2015, 2038, 2035, 2045, 2041, 2036, 2040, 2042, 2065, 2022, 2038, 1986 2030, 2040, 2023, 2018, 2036, 2040,
1953, 1958, 1987 1948, 1938, 1914, 1889, 1878, 1900, 1956, 1900, 1956, 1900 1900, 1966, 1916, 1920,
1905 1918
-2.0 0 7.4 1900 10.5 11.8 1900, 1838 1884 11.9 12.1 1860 1836 13.4 1836 15.8 1932, 1920, 1900 16.0 1884, 1844 17.0 1926, 1908, 1894 17.1 19.3 19.7 1869 21.9 1942 22.1 1900, 1884 22.2 1910 22.6 1932, 1906 23.0 1956, 1932, 1924, 1900 24.3 1891, 1872 24.5 1926, 1887 24.7 1954, 1922, 1900 26.8 1976, 1924 27.5
25.7 26.7 30.4 32.0 32.6 32.7 32.8 33.4 34.6 34.7 35.2 35.2 36.2 36.6 37.7 37.8 37.8 38.0 38.2 38.9 39.0 39.1 40.1 40.5
2038, 2040, 2038, 2020, 2036,
1954, 1926 1952, 1954, 1942,
1914, 1900, 1884 30.1 30.1 1914, 1900, 1884 32.1 32.4 1920, 1900 1920, 1900 32.9
41.8 41.8 42.8 42.9 43.2
Infrared spectra were recorded in THF solution.
system increase, it should be possible to make good estimates for stabilities of a number of complexes and proposed intermediates. Chelating Nitrogen Donors. As mentioned before, the heat of binding of bpy agrees within experimental error with that of 2 mol of pyridine. The heat of binding of the tertiary aliphatic amine TMED is only slightly more negative than for the less basic pyridine system. It was surprising to find that the 1,lOphenanthroline complex was some 5 kcal/mol more stable than the corresponding bpy complex. Two rationalizations come to mind. The first is that the phen ligand has increased ability to function as a ‘IT acceptor due to its more delocalized ‘IT orbitals. The second is that the configuration of free bpy has the two pyridines perpendicular to each other. Displacement of bound bpy by phen will allow relaxation to the preferred geometry of the free ligand as shown in eq 14. Molecular mechanics calcu-
[MY
2 PPhzCHzCHzASPhz
+
+
(13)
ASPh,
C‘
/MO(co)4
ASPh2
bond -AH strength
l u
Ph2PCH2CHzPPhz
+ PhZAsCHzCH, AsPh,
of a r p h o ~ M o ( C 0 produced. )~ As thermochemical data for this (12) Bruno, J. W.; Marks, T. J.; Morss, L. R. J. Am. Chem. SOC.1983,105, 6824. (13) Grim, S. 0.; Briggs, W. L.; Barth, R. C.; Tolman, C. A,; Jesson, J. P. Inorg. Chem. 1914, 13, 1095. (14) (a) Cheung, K. K.; Lai, T.F.; Mok, K. S. J . Chem. SOC.A 1971, 1644. (b) Favas, M. C.; Kepert, D. L.; Skelton, B. W.; White, A. H. J . Chem. SOC., Dalton Trans. 1980, 447. (15) Page, M. I . Angew. Chem., Int. Ed. Engl. 1977, 16, 449.
(14)
lations16 appear to support this second explanation. The complex ~~~~~
~
~
(16) Gawley, R., private communication
Inorg. Chem. 1988, 27, 85-92 of dimethylphenanthroline is 1.2 kcal/mol less stable than the unsubstituted phenanthroline ligand, presumably due to steric repulsion between the methyl groups and coordinated carbon monoxide ligands. Cyclooctadiene versus Norbornadiene. As discussed in this section, substitution of norbornadiene for cyclooctadiene as shown in eq 15 is exothermic by about 2 kcal/mol. This is opposite to
W
W
4
(15)
the heat of binding to Pd(II), where cyclooctadiene was 0.3 kcal/mol more stable.” The reasons for these relatively small differences are not clear. Conclusion Data for monodentate and bidentate ligands are combined in Table V, which also includes IR spectral data. In general, the stabilities of L2Mo(CO), complexes are similar to those in our earlier work on L,Mo(CO),. The importance of steric factors is reduced, as expected. The changing electronic nature of the metal center may also play a role in influencing complex stability. For bidentate ligands forming metallacycles, four-membered rings are less stable than five-membered rings by 8 kcal/mol. Six- and seven-membered rings are slightly less stable than four-membered (17) Partenheimer, W. Inorg. Chem. 1972, 11, 743.
85
rings (1-2 kcal/mol). In the absence of ring strain, or special steric effects, heats of reaction can be estimated to fair accuracy on the basis of data for related complexes. Additional work in progress is aimed at extending those data in order to generate a complete picture of the factors controlling stability in these and other organomolybdenum systems. Acknowledgment. Support of this work by the National Science Foundation (Grant No. CHE-8618753) is gratefully acknowledged. Registry No. dppe, 1663-45-2; COD, 111-78-4; NBD, 121-46-0;bpy, 366-18-7; dpae, 4431-24-7; tmeda, 110-18-9; Mephen, 484-1 1-7; dppm, 2071-20-7; phen, 66-71-7; arphos, 23582-06-1; dppb, 7688-25-7; dppp, 6737-42-4; dmpm, 64065-08-3; dppbz, 13991-08-7; dmpe, 23936-60-9; py, 110-86-1; (PC13)2Mo(C0)4, 16244-51-2; ( P ~ ~ A s ) ~ M o ( C16742O)~, 97-5; [P(OM~),],MO(CO)~,15631-22-8; ( ~ y ) ~ M o ( C 016742-99-7; )~, (Ph,P),Mo(CO),, 16742-93-1; ( E ~ ~ A s ) ~ M o ( C O11) ~1265-67-9; , [P(OPh),],Mo(CO),, 59599-01-8; (PPh,Me),Mo(CO),, 37438-49-6; ( P P ~ M ~ , ) , M o ( C O ) ~24554-47-0; , ( P - ~ - B U , ) , M O ( C ~ 16244-54-5; )~, (CyNC),Mo(CO),, 15227-72-2; (PEt,),Mo(CO),, 19217-80-2; (PMe3),Mo(C0),, 16027-45-5; (COD)MO(CO)~,12109-74-9; (bpy)Mo(CO),, 15668-64-1; (dpae)Mo(C0)4, 38536-63-9; (dppe)M~(CO)~, 14971-45-0; (tmeda)Mo(CO),, 23301-98-6; ( d p p m ) M ~ ( C O ) ~26743, 8 1-7; ( p h e n ) M ~ ( C O ) ~15740-78-0; , (arpho~)Mo(CO)~,53557-42-9; (dppb)Mo(CO),, 15553-69-2; (dppp)Mo(CO),, 15553-68-1; (dppe)Mo(CO),, 15444-66-3; (dmpm)Mo(CO),, 90624-09-2; (dppbz)Mo(C0)4, 111189-30-1; (dmpe)Mo(CO),, 40544-97-6; (NBD)Mo(CO),, 1214637-1; Mo(CO),, 13939-06-5; Ph,As, 603-32-7; Ph3P, 603-35-0; Et,As, 617-75-4; P(OPh),, 101-02-0; PPh,Me, 1486-28-8; CO, 630-08-0; PPhMe,, 672-66-2; P(OMe),, 594-09-2; P-n-Bu,, 998-40-3; CyNC, 931-53-3; PEt,, 554-70-1; PMe,, 594-09-2; PCl,, 7719-12-2; P, 772314-0; Mo, 7439-98-7.
Contribution from the Centro di Chimica e Tecnologia dei Composti Metallorganici degli Elementi di Transizione del C N R , Istituto di Chimica Industriale, Facolti di Ingegneria, Universiti di Padova, 35 100 Padova, Italy, Dipartimento di Chimica “G. Ciamician”, Universiti di Bologna, 401 26 Bologna, Italy, and Department of Chemistry, Iowa State University, Ames, Iowa 5001 1
Transition-Metal-Promoted Cyclization Reactions of Isocyanide Ligands. Synthesis of Cyclic Aminooxycarbene Complexes of Platinum(I1) and X-ray Structure of
trans -{ (PPh3)2Pt[CN(C6H4-p -Me)CH2CH20]BrJBF4 Rino A. Michelin,*la Livio Zanotto,la Dario Braga,* l b Piera Sabatino,lb and Robert J. Angelici*“ Received February 19, 1987 Electrophilic isocyanide CNR ligands in cationic Pt(I1) complexes of the type trans-[ (PR’,),Pt(CNR)Cl]BF, (I) (PR’, = PPh3, PMe2Ph; R = p-MeOC6H4, p-MeCaH4,p-N02C6H4, Me, C6Hll) are converted to the corresponding 5-membered cyclic aminooxycarbene derivatives rrans-((PR’3)2Pt[CN(R)CH2CH20]X)BF4 (11) (X = C1, Br) by reaction in T H F with 2-bromoethanol in the presence of n-BuLi. These reactions are likely to proceed by nucleophilic attack of the alkoxide on the isocyanide carbon atom to give an imidoyl intermediate, which cyclizes intramolecularly to yield the carbene products. The less sterically hindered CNMe and aryl isocyanide ligands in I are converted in a few minutes to the final products I1 in ca. 70-90% yield; the more bulky does not react CNC6HlIderivative gives only a 25% yield. The t-BuNC ligand in the complex ?rans-[(PPh3),Pt(CN-t-Bu)C1]BF4 at all, nor does p-MeOC6H4NC in trans-[(PCy3)2Pt(CNC6H4-~oMe)cl]BF4 with bulky PCy, ligands. Treatment of cisCl2Pt(CNC6H4-p-0Me),with 2 equiv of 2-bromoethoxide gave the bis(aminooxycarbene) Br2Pt[CN(C6H4-p-OMe)CH2CH20]2. The Pt(II)-cyclic aminooxycarbenes I1 were characterized by their elemental analysis and IR, ‘HNMR, and NMR spectra. An X-ray-determined structure of ~ ~ ~ ~ S - ( ( P P ~ ~ ) ~ P ~ [ & N ( C ~ H ~ - ~ - space M ~ ) group CH~C P2,/a, H ~ Oa]=B 12.175 ~ } B F(2) ~ ,A, b = 26.137 (3) A, c = 13.274 (4) A, @ = 91.61 (2)O, and 2 = 4, was refined to R = 0.039 (R, = 0.044) for 5325 independent reflections. The coordination geometry around the Pt(I1) atom is square planar with the carbene ligand perpendicular to the plane. The cyclic aminooxycarbene ligand is planar with C(sp2)-N and C(sp2)-0 bond distances of 1.30 (1) and 1.33 (1) A, indicating significant *-bonding between the nitrogen, oxygen, and carbene carbon.
Introduction Cyclization reactions of electrophilic metal-coordinated isocyanide ligands leading to heterocyclic carbene complexes have been accomplished by different synthetic strategies (Scheme I).zd (1) (a) CNR, Padova. (b) UniversitH di Bologna. (c) Iowa State University. (2) Bartel, K.; Fehlhammer, W. P. Angew. Chem., Inf. Ed. Engl. 1974,13, 599. Plaia, U.; Fehlhammer, W. P. J. Am. Chem. SOC.1985, 107, 2171. Fehlhammer, W. P.; Bartel, K.; Plaia, U.; Volkl, A.; Liu, A. T. Chem. Ber. 1985, 118, 2235. 0020-1669/88/ 1327-0085$01.50/0
For instance, paths a and b of Scheme 1 take advantage of the reactivity of suitably functionalized isocyanides which undergo (3) (a) Facchin, G.; Campostrini, R.; Michelin, R. A. J . Organomet. Chem. 1985, 294, C21. (b) Michelin, R. A.; Facchin, G . ;Braga, D.; Sabatino, P. Organometallics, 1986, 5, 2265. (4) Fehlhammer, W. P.; Bartel, K.; Petri, W. J . Organomet. Chem. 1975,
87, c34.
( 5 ) Grundy, K. R.; Roper, W. R. J . Organomet. Chem. 1975, 91, C61. (6) Fuchita, Y . ;Hidaka, K.; Morinaga, S.; Hiraki, K. Bull. Chem. SOC.Jpn. 1981, 54, 800.
0 1988 American Chemical Society
